Mechanical Evaluation of Cross-Link Designs in Rigid Pedicle Screw Systems

Abstract

Study Design. This study was designed to evaluate the biomechanical performance of 5 different cross-link brands to determine which design characteristics are biomechanically desirable. Methods. The Cotrel-Dubousset, Isola, Puno Winter Byrd, Rogozinski, and Texas Scottish Rite Hospital sys tems were assembled to vertebral models according to the manufacturer's specifications. Three constructs were tested for each brand of instrumentation: without cross links, with one cross-link, and with two cross-links. Four modes of loading: axial, torsional, flexion-extension, and lateral-flexion were used. Load-displacement curves were plotted. The stiffness was calculated from the slope of these curves. Objectives. Five different rigid pedicle screw systems were tested to determine: 1) what are the characteristics of cross-link design that are most effective in limiting torsional motion; 2) whether two cross-links are more effective than one; and 3) whether cross-linkage in creases the construct stiffness in lateral bending. Summary of Background Data. Cross-linkage has been shown to increase the torsional stiffness of rod and screw constructs. Increased construct stiffness has been correlated with higher fusion rates. Results. Increases in axial, flexion-extension, or lat eral-flexion stiffness, with the addition of one or two cross-links, were not statistically significant. In torsional loading, increases in stiffness within brands were statis tically significant in every case. The average increase was 44% with one added cross-link and 26% with two. The magnitude of the increase in torsional stiffness was compared with the cross-sectional area of the respec tive cross-link. Greater stiffness correlated with larger cross-sectional area (r = 0.81 for one cross-link, and r = 0.60 for two). Conclusion. The use of cross-linkage in spinal fusion increases torsional stiffness in pedicle screw and hook constructs. This study 1) confirmed the effectiveness of cross-linkage in limiting torsional motion and showed the superiority of two cross-links to one cross-link in limiting torsional motion, 2) showed that increase of torsional stiffness of a cross-linked construct is propor tional to the cross-sectional area of the cross-link, and 3) demonstrated that cross-links do not increases stiffness in the lateral flexion mode.